We live in a world where large quantities of data are regularly collected about people, institutions, and social structures. This course will demonstrate how quantitative analysis techniques can be used to leverage this data and answer complex questions about the social world, including:
- Why are some people are more at risk of crime than others?
- What explains differences in life expectancy between countries?
- Do gender inequalities persist in the workplace?
This course investigates the underlying principles and uses of statistical models and not on mathematical and statistical theory. It will give you a solid empirical grounding to be able to critically evaluate the findings from a wide range of quantitative social science research
You will get hands-on experience of estimating a number of different statistical models in R, engaging with important issues including how to select an appropriate model, assessing the adequacy of a fitted model (in comparison to alternative models), and the statistical and substantive interpretation of the results.
On successful completion of this course, you will be able to:
- Have a critical awareness of the rationale and terminology of statistical modelling (C)
- Engage with existing quantitative research, highlighting its key strengths and weaknesses (C and K)
- Have a comprehensive understanding of the logic of model development and testing (C and K)
- Develop multiple regression, logistic regression, multinomial logistic and poisson regression models and critically evaluate the results (P and T)
- Clearly tabulate and present the results of regression outputs (P and T)
This course elaborates on quantitative approaches to social science, combining this with practical model building experience and critique using R.
Indicative content includes:
- Designing and building statistical models to answer social science questions
- The general linear model
- Operationalising concepts and selecting variables
- Interpreting results and finding the narrative.
Practical workshops will provide you with experience of:
- Linear regression
- Logistic regression
- Multinomial regression
- Poisson regression
- Interaction effects and nonlinear relationships
- Model fit and diagnostics
- Missing data adjustments.
Learning and teaching methods
- Practical workshops in R
- Group discussion and feedback
Ian Brunton-SmithSee profile
Allum, N., Besley, J., Gomez, L., and Brunton-Smith, I. (2018) Disparities in science literacy. Science, 360 (6391), pp.861-82.
There are no formal entry requirements for this course.
You should have some knowledge of regression.
Fees and funding
Price per person:
£595Government and commercial sector applicants
£495Education and charitable sector applicants
Terms and conditions
When you accept an offer of a place at the University of Surrey, you are agreeing to comply with our policies and regulations and our terms and conditions. You are also confirming you have read and understood the University's prospective student privacy notice.
Further details of our terms and conditions will follow.
This online prospectus has been prepared and published in advance of the commencement of the course. The University of Surrey has used its reasonable efforts to ensure that the information is accurate at the time of publishing, but changes (for example to course content or additional costs) may occur given the interval between publishing and commencement of the course. It is therefore very important to check this website for any updates before you apply for a course with us. Read more.